Google's August 2026 Spam Update: What It Targets

On August 19, an owner forwarded me an email her Search Engine Optimization (SEO) vendor had just sent to its whole client list: a form letter reassuring everyone their rankings were "fully protected" against "the recent Google update." She wanted to know which update. Fair question, because the day before, Google actually had shipped one — and it isn't the kind of thing a form letter protects anyone from.

That's usually how this goes. Google ships something real and specific. Somewhere, a vendor turns it into a generic reassurance email, because "don't worry, we've got you" is easier to write than "here's exactly what changed and whether it applies to you." Let's do the second one.

What this update actually targets

Google's August 2026 spam update enforces its existing spam policies, not new ones. Two categories are doing the work here: cloaking and scaled content abuse.

Cloaking is showing search engines one version of a page and visitors a different one — rare enough that most small business owners will never touch it, deliberately or otherwise. Scaled content abuse is the one that actually matters for this audience. In Google's own written spam policies, it isn't about whether content was written by a person or generated with AI assistance — it's about purpose and volume. Pages produced in bulk, on a template, primarily to occupy search results rather than to say something a specific reader needed, qualify as scaled content abuse regardless of who or what typed the words. A hundred nearly identical "[Service] in [City]" pages with the city name swapped out and nothing else genuinely different is the textbook example, and it's also a product a lot of cheap SEO packages have been selling for years, AI tools or not.

This isn't only a traditional-search concern, either. Google's spam policies now state explicitly that manipulating AI Overview and AI Mode responses counts as spam under the same rules as manipulating a traditional results page. Different surface, identical logic: the discipline of earning an AI citation instead of manufacturing one runs on the same principle a spam-update-proof site already needs — say something specific enough to be worth citing, rather than producing volume and hoping some of it sticks.

Worth being precise about what this update did not target, because vendors love to blame the thing that isn't the actual cause: link spam and site reputation abuse — both targets of earlier Google spam updates — are explicitly excluded from this rollout. If your traffic dropped after August 18 and someone's first move is a backlink audit, they're solving a problem this update didn't create.

Why this one is worth actually reading, unlike most

Most of what moves a small business site's rankings never gets a name, a date, or an announcement at all — Google has said outright that it makes constant smaller changes it doesn't bother disclosing, which is exactly why so many ranking slides turn into unsolvable mysteries. This is the opposite kind of event: named, dated, logged on Google's own Search Status Dashboard, with a clearly stated target. That's rare enough to be worth the five minutes it takes to actually understand it, instead of filing it under "another algorithm thing" and moving on.

It's also aimed, whether Google frames it this way or not, at a specific sales pitch a lot of small business owners have already bought: the SEO vendor who measures their own value in publishing volume. "We'll post 30 blog articles a month" or "we'll build you a page for every city in your service area" sounds like effort, and it photographs well in a monthly report. Content that exists to fill a report rather than answer a real question was already a losing bet before this update — most of it earns zero organic traffic regardless of what's chasing it. What changed on August 18 is that "primarily built to rank, not to help" now has a specific enforcement action attached to it, on a specific date, instead of just quietly underperforming forever.

The mechanics: how to tell if this actually touches you

A few concrete things to check, in order:

1. Look at the actual date. Pull up Search Console's Performance report and check whether impressions or clicks dropped specifically starting around August 18–21, not sometime vaguely "this month." A drop with a clean start date lines up with an update. A slow multi-week slide with no clear starting point is more likely the ordinary, unannounced kind of drift.
2. Audit your own content for the pattern, not the label. Scan your published pages for the actual shape of scaled content abuse: near-duplicate pages differing mainly by a swapped city or keyword, thin pages with no specific detail about your business, anything published in a sudden batch rather than over time. AI involvement isn't the flag — sameness and volume-for-its-own-sake are.
3. Ask your content vendor a direct question. Not "were we affected by the update" — that invites a form letter. Ask: "Does a person read and edit each page before it publishes, and does each one say something a competitor's equivalent page doesn't?" A vague answer is itself useful information.
4. Rule out the wrong diagnosis. If a drop lines up with this update and someone reaches for a backlink cleanup or disavow file, that's solving for the previous update, not this one. This round didn't touch links at all.

What won't help

• Deleting every page that used AI assistance. The update doesn't target the tool. It targets pages that exist primarily to rank rather than to help someone — which plenty of AI-assisted content, reviewed and specific, does not do.
• Waiting on a vendor's mass reassurance email. A form letter sent to an entire client list the day after an update ships is a marketing response, not a diagnosis of your specific site.
• Running a link audit for this one. Link spam and site reputation abuse weren't part of this rollout. That's a different problem, from a different update.
• Assuming every dip this month is this update. Plenty of ranking movement still comes from the quiet, unnamed changes Google runs constantly in the background — worth ruling that in or out separately before deciding which fix applies.

If your traffic did something specific and unexplained right around mid-August and you genuinely don't know whether it's this update, a template problem from an old vendor, or something else entirely, that's exactly the kind of question the Fix path exists to answer with an actual look at the site instead of a form letter.

FAQ

What is Google's August 2026 spam update?

It's the third spam-focused algorithm update Google shipped in 2026, following updates in March and June. Released August 18 and rolled out globally across all languages over roughly two and a half days, it enforces Google's existing spam policies against cloaking and scaled content abuse — it isn't a new policy, just active enforcement of a standing one.

Does this update target AI-generated content specifically?

No. Google's stated position is that content violates its spam policies based on its purpose and volume, not on whether AI helped write it. A page written entirely by a person can qualify as scaled content abuse if it's one of hundreds of near-identical pages built mainly to occupy search results. AI-assisted content that's edited, specific, and actually useful isn't the target.

Did this update also target spammy backlinks?

No — link spam and site reputation abuse, which some earlier Google spam updates have gone after, are explicitly not part of this rollout. If a drop coincides with this update, a backlink cleanup is very likely solving the wrong problem.

How do I know if my site was hit by this specific update?

Check Search Console for a traffic or impression drop with a clean start date around August 18–21, rather than a vague slide over weeks. Then look at your own content for the actual pattern this update targets: near-duplicate pages, thin location or service pages published in a batch, and content that reads like it exists to rank rather than to answer something specific.

I bought a package with dozens of AI-written city pages. Should I be worried?

It depends on whether those pages are genuinely distinct and useful or interchangeable with the city name swapped. The test isn't how they were written — it's whether someone searching your city specifically would learn something a competitor's identical page doesn't say. If the honest answer is no, that's the pattern this update enforces against, regardless of who wrote it.

Is a Google spam update the same as a core update?

No. Core updates broadly reassess how Google evaluates page quality and relevance across the board. Spam updates are narrower — they specifically target sites violating Google's written spam policies, like cloaking or scaled content abuse, and don't otherwise touch how quality content gets ranked.

What should I actually do if I think this update affected me?

Start by confirming the timing lines up, then honestly assess whether any of your published content matches the pattern — mass-produced, near-duplicate, built to occupy rankings rather than help a specific visitor. If it does, the fix is consolidating or substantively rewriting those pages with real, specific detail, not waiting for the next update to pass judgment differently.